Performance evaluation of aodv and olsr in vanet under realistic mobility


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Performance evaluation of aodv and olsr in vanet under realistic mobility

  1. 1. INTERNATIONAL JOURNAL OF ELECTRONICS AND International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEMECOMMUNICATION ENGINEERING & TECHNOLOGY (IJECET)ISSN 0976 – 6464(Print)ISSN 0976 – 6472(Online)Volume 4, Issue 2, March – April, 2013, pp. 58-71 IJECET© IAEME: Impact Factor (2013): 5.8896 (Calculated by GISI) © PERFORMANCE EVALUATION OF AODV AND OLSR IN VANET UNDER REALISTIC MOBILITY PATTERN R.Boopathi, M.Tech, R.VishnuPriya. M.E. V.R.S. College of technology, Arasur, Villupuram. Tamilnadu, India ABSTRACT A Vehicular Ad Hoc Network (VANET) is an instance of MANETs that establishes wireless connections between cars. In VANETs, routing protocols and other techniques must be adapted to vehicular-specific capabilities and requirements. As many previous works have shown, routing performance is greatly dependent to the availability and stability of wireless links, which makes it a crucial parameter that should not be neglected in order to obtain accurate performance measurements in VANETs. Although routing protocols have already been analyzed and compared in the past, simulations and comparisons have almost always been done considering random motions. But could we assess that those results hold if performed using realistic urban vehicular motion patterns? In this paper, we evaluate AODV and OLSR performance in realistic urban scenarios. We study those protocols under varying metrics such as node mobility and vehicle density, and with varying traffic rates. We show that clustering effects created by cars aggregating at intersections have remarkable impacts on evaluation and performance metrics. Our objective is to provide a qualitative assessment of the applicability of the protocols in different vehicular scenarios. Index Terms — Urban Environment, Realistic Vehicular Mobility Models, OLSR, AODV, Performance. 1. INTRODUCTION Vehicular Ad-hoc Networks (VANETs) represent a rapidly emerging, particularly challenging class of Mobile AdHoc Networks (MANETs). VANETs are distributed, self organizing communication networks built up by moving vehicles, and are thus characterized by very high node mobility and limited degrees of freedom in the mobility patterns. Hence, ad hoc routing protocols must adapt continuously tothese unreliable conditions, whence the growing effort in the development of communication protocols which are specific to vehicular networks. 58
  2. 2. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEME One of the critical aspects when evaluating routing protocols for VANETs is theemployment of mobility models that reflectas closely as possible the real behavior ofvehicular traffic. This notwithstanding, using simple random-pattern, graph constrainedmobility models is a common practice among researchers working on VANETs. There is noneed to say that such models cannot describe vehicular mobility in a realistic way, since theyignore the peculiar aspects of vehicular traffic, such as cars acceleration and deceleration inpresence of nearby vehicles, queuing at roads intersections, traffic bursts caused by trafficlights, and traffic congestion or traffic jams. All these situations greatly affect the networkperformance, since they act on network connectivity, and this makes vehicular specificperformance evaluations fundamental whenstudying routing protocols for VANETs. Initialworks [2], [5] on performance evaluation considered simple pseudorandom motion patternsand lacked any interaction between cars, generally referred as micro-mobility. Following therecent interest in realistic mobility models for VANETs, new studies appeared onperformance evaluations of VANETs in urban traffic or highway traffic conditions [13], [14].However, their models were quite limited, notably the macro-model, which also limited thescope of their results. In this paper, our objective is to evaluate AODV and OLSR in realistic urban trafficenvironment. In order to model realistic vehicular motion patterns, we make use of theVehicular Mobility Model (VMM), which is part on the VanetMobiSimtool we previouslydeveloped (see [8]). This model is ableto closely reflect spatial and temporal correlationsbetweencars, and between cars and urban obstacles. Notably, the tool illustrates clusteringeffects obtained at intersection, also is more commonly called traffic jam, or drastic speeddecays. Accordingly, it becomes possible to more realistically evaluatead hoc routingperformances for vehicular networks. Weconfigure VMM to model urban environment thenevaluatethe performance of AODV and OLSR in terms of (i) Packet Delivery Ratio (PDR)(ii) Control Traffic Overhead (RO), (iii) Delay, and (iv) Number of Hops. We test AODV andOLSR in three different conditions (i) variable velocity (ii) variable density (iii) variable datatraffic rate. We show first that the clustering effect obtained at intersection has a major effecton the effective average velocity during the simulation. We then illustrate how OLSR is ableto outperform AODV in any condition and for almost all metrics. The rest of the paper is organized as follows: In Section II, we shortly provide somerelated work in the performance evaluation field, while in Section III, we briefly depictOLSR and AODV protocols. Section IV presents the Vehicular Mobility Model (VMM) weused in this paper to model urban motion patterns, while Section V discusses the scenariocharacteristics and the simulation results. Finally, in Section VI, we draw some conclusionremarks and outline some future works.2. RELATED WORK Several studies have been published comparing the performance of routing protocolsusing different mobility models or performance metrics. One of the first comprehensivestudieswas done by the Monarch project [2]. This study compared AODV, DSDV, DSR andTORA and introduced some standard metrics that were then used in further studies ofwireless routing protocols. A paper by Das et al. [5] compared a larger number of protocols.However, link level details and MAC interference are not modeled. Another study [9]compared the same protocols as the work by Broch et al. [2], yet for specific scenarios as theauthors understood that random mobility would not correctly model realistic networkbehaviors, and consequently the performance of the protocols tested. Globally, all of thesepapers concluded that reactive routing protocols perform better than proactive routingprotocols. Although that the proactive OLSR protocol has been developedin 2002, very few 59
  3. 3. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEMEstudies compared it with other ad hoc network protocols. Clausen et al. [15] evaluatedAODV, DSR and OLSR in varying network conditions (node mobility, network density) andwith varying traffic conditions (TCP, UDP). They showed that unlike previous studies, OLSRperforms comparatively to the reactive protocols. Following the developments started withscenarios-based testing, it also became obvious that, as scenarios were able toalter protocolperformances, so would realistic node-to-node or node-to-environment correlations. Thisapproach became recently more exciting as VANETs attracted more attention, anda newwave of vehicles-specific models appeared. The most comprehensive studies have beenperformed by the Fleet net project [18]. In a first study [13], authors compared AODV, DSR,FSR and TORA on highway scenarios, while [14] compared the same protocols in city trafficscenarios. They found for example that AODV and FSR are the two best suited protocols,and that TORA or DSR are completely unsuitable for VANET. Another study [12] compareda position-based routing protocol (LORA) with the two non-position-based protocols AODVand DSR. Their conclusions are that, although AODV and DSR perform almost equally wellunder vehicular mobility, the location-based routing schema provides excellent performance.A similar results has been reached by members of the NoW project [19], which was theirmajor justification for the design of Position-based forwarding techniques. However, to thebest of our knowledge, no performance evaluations have been conducted between OLSR andother routing protocols under realistic urban traffic configurations.3. BACKGROUND For our performance comparison study, we picked up twoad hoc routing protocolsthat reached the IETF RFC stage, theon-demand AODV protocol (RFC[3561] [11]), and thetable driven OLSR protocol (RFC[3626] [3]). We shortly address both protocols in the rest ofthis section. For a more detailed description, the reader is referred to the respective RFCs.A. AODV (Ad-hoc On-demand Distance Vector) In AODV, when a source node has data traffic to send toa destination node, it firstinitiates a route discovery process. In this process, the source node broadcasts a RouteRequest (RREQ) packet. Neighbor nodes which do not know an activeroute for the requesteddestination node forward the packet totheir neighbors until an active route is found or themaximum number of hops is reached. When an intermediate node knowsan active route tothe requested destination node, it sends a Route Reply (RREP) packet back to source node inunicastmode. Eventually, the source node receives the RREP packetand opens the route.B. OLSR (Optimized Link State Routing) In OLSR, each node periodically constructs and maintains the set of neighbors thatcan be reached in 1-hop and 2-hops. Based on this, the dedicated MPR algorithm minimizesthe number of active relays needed to cover all 2-hops neighbors. Such relays are calledMulti-Point Relays (MPR). A node forwards a packet if and only if it has been elected asMPR by the sender node. In order to construct and maintain its routing tables, OLSRperiodically transmit link state information over the MPR backbone. Upon convergence, anactive route is created at each node to reach any destination node in thenetwork.4. VEHICULAR MOBILITY MODEL As depicted in [4], a mobility model clearly affects the simulation results. Thus, sincesimple models like the Random Way point mobility model do not consider vehicles’ specificmotion patterns, they cannot be applied to simulation of vehicularnetworks. Accordingly, wedeveloped in [8] a new realistic mobility model, called Vehicular Mobility Model (VMM) that is 60
  4. 4. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEMEcompliant with the principles of the general frame work for mobility models generation describedin [6], and capable of modeling detailed vehicular movements in different trafficconditions.Following the general classification proposed by [7], VMM contains a microscopic and amacroscopic component:A. Macro-Mobility The macro-model is represented by a graph where verticesand edges represent,respectively, junction and road elements. As proposed by [10], a good solution to randomlygenerate graphs on a particular simulation area is Voronoi tessellations based on distributedpoints over the simulation area which represent obstacles (e.g., buildings). Accordingly, weobtain aplanar graph representing a set of urban roads, intersections and obstacles. Then, in orderto increase the realism, as denseareas such as city centers have a larger number of obstacleswhich in turn increase the number of Voronoi domains, the model generates clusters of obstacleswith different densities, eventually creating clusters of Voronoi domains. Figure 1(a) presents arandom topological map with uniformly spread obstacles, while Figure 1(b) depicts topologicalmap considering three different types of clusters with different obstacle densities. (a)Uniform Topology (b) Cluster Topology Fig. 1. Illustration of the random topology generation In order to model the typical vehicular motion patterns, the objective is also to create arelationship between the topological map and the traffic generator that could go beyond thesimple constrained motions induced by graph-based mobility. Accordingly, the macro-model firstoffers the possibility to separate single flows roads, as well as to increase the number of lanes perroad. Then, as the traffic generator needs to act when reaching an intersection, the urban topologyis also enhanced by traffic signs. According to the model’s configuration, traffic lights or stopsigns may be added, depending on the type of intersection.B. Micro-Mobility When considering micro-mobility, one should look at the driver’s point of view. When adriver approaches an intersection, it should slow down then act according to the traffic signs ortraffic lights he or she reads, and to the presence of other cars approaching the sameintersection.To obtain a similar behavior, the existing Intelligent Driver Model [16] is extended toderive the Advanced Intelligent Driver Model (AIDM) supporting intersection management.Tothis end, deceleration and acceleration models inspired by the Akcelik’sacceleration/deceleration model [1] are added in proximity of road intersections, so that vehiclesapproaching a traffic light or a cross road reduce their speeds or stop. Included are also a set ofrules describing the actions taken by drivers at intersections depending on the class of trafficsigns, the state of traffic lights and other vehicles currently inside the intersection or waiting fortheir turns. Finally, avehicle overtaking model has also been included in order to allow vehicles tochange lane and overtake each others. We chose the Minimizing Overall Braking decelerationsInduced by Lane changes (MOBIL) [17] model as the lane changing model, due to its implicitcompatibility with the AIDM. 61
  5. 5. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEME5. PERFORMANCE EVALUATION In order to evaluate the performance of the routing protocols described above, we usedthe open source network simulatorns-2 in its version 2.27 as it is widely used for research inmobile ad hoc networks. We provide first a description of the scenarios characteristics and thendescribe the results we obtained.A. Scenario Characteristics In this paper, we consider squared urban areas of 1000x1000m constituted of threedifferent cluster categories: downtown, residential and suburban. The different obstacle densitiesfor these three categories are summarized in Table II (b). Vehicles are able to move freely on theurban graph respecting roads and intersection rules, more specifically, speed limitations andstops. Vehicles are able to communicate with each other’s using the IEEE 802.11 DCF MAClayer. The radio transmission range as been deliberately over-evaluatedand set to 250m forVANETs as we wanted to avoid biased performance evaluations due to disconnected networks.The simulation parameters are given in Table I. We test each protocol with a spatial model withstop signs only, and with 30% of traffic lights and 70% of stop signs, as we also want to evaluatethe effect of traffic lights in urban areas.Vehicles are randomly positioned on intersections. Then,each vehicle samples a desired speed and a target destination. After that, it computes the shortestpath to reach it, taking into account single flow roads. Eventually, the vehicle moves andaccelerates to reach a desired velocity according to streets regulations. When a car moves nearother vehicles, it tries to overtake them if the road includes multiple lanes. If it cannot overtake, itdecelerates to avoid the impact. When a car is approaching an intersection, it first acquires thestateof the traffic sign. If it is a stop sign or if the light is red, it decelerates and stops. If it is agreen traffic light, it slightly reduces its speed and proceeds to the intersection. Attargetdestination, the car decelerates and stops, then samples a new destination. The differentparameters for the micro-model aregiven in Table II(a)We finally decomposed our performanceanalysis in three different scenarios, where we fixed the parameters accordingto Table III. In thefirst scenario, we want to see the influence of mobility, where as in the second scenario, we areinterested in the data traffic rate, and finally, in the last scenario, the objective aims at observingthe effect of the network density. Each point is the average of 10 samples, while the errorbarsrepresent a 95% confidence interval. TABLE I SIMULATION PARAMETERS 62
  6. 6. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEME TABLE II VEHICULAR MOBILITY MODEL PARAMETERSB. Simulation Results We measured several significant metrics for MANETs routing:• Packet Delivery Ratio (PDR)– It is the ratio between the number of packets delivered to the receiver and the number of packets sent by the source.• Routing Overhead (RO)– It represents the number of routing bytes required by the routing protocols to construct and maintain its routes.• Delay– It measures the average end-to-end transmission delay by taking into account only the correctly received packets.• Hops– It provides an expected data route length. We see in Fig. 2 that the average velocity does not have any effect on the PDR, which is a strange results as mobility is a common metric in performance evaluation, and previous results have shown that both protocols were sensitive to it. We also obtained similar behaviors for other performance metrics, but did not include them for the lack of space. Actually, the explanation for this behavior comes from the micro-model Table III Simulation Scenariosand its interaction with the spatial environment. Indeed, when modeling smooth transitionsand realistic interaction with urban traffic regulations, a fixed initial velocity does not makeany sense. Instead, we define an average desired velocity adriver aims at reaching with a 63
  7. 7. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEMEsmooth acceleration. However, this desired velocity is subject to speed limitations that cannot beexceeded, or to any obstacle that either reduces or even stops the car. Accordingly, there is noguarantee that this velocity can even be reached during the simulation. Fig. 2. Packet Delivery Ratio (PDR) as a function of Average Velocity In order to illustrate this effect, we show in Fig. 3(a)and Fig. 3(b) the speed decay fromthe desired speed thatvehicles experience in our scenarios. As we can see, thereis a drastic decayfor either a varying density or varying desired velocity. The question we may ask is what isthemain limiting factor that leads to this effect? We can see inFig. 3(c) that one on the parametersis cars acceleration (resp.deceleration). Actually, this should not be strange as when we observeurban traffic; smooth transitions are major criteriafor traffic jams (even without any obstacles).On the same figure, we also see that as the speed decay stabilizes for large accelerations, thedecay actually becomes dependent to the distance between two intersections, which is a secondparameter which influences cars velocity in our model. Finally, we can see, on Fig. 3(d), the nonuniform distribution of vehicles on the simulation area, illustrating yet another specificity ofrealistic mobility modeling creating this effect. The major conclusion is that pure mobility asdefined by previous works cannot be used as an evaluation metric for vehicular ad hoc networks.We should rather define new metrics as acceleration/deceleration factors, or distance between twointersections. Fig. 3. Illustration of Vehicular-specific Motion Patterns 64
  8. 8. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEME Figure 4 shows the average PDR against the CBR through put. The first observation wecan make is that neither AODVnor OLSR clearly outperforms the other. The PDR variationbetween both protocols never goes above 10%. This small variation comes from AODV andOLSR core functionalities. Indeed, in our scenarios, we are increasing the data traffic rate, yetkeeping the number of CBR source constant. For small datarates, OLSR performs better due tothe fact that all routes arecomputed at no extra cost, while AODV must initiate several routediscovery processes. When the rate of route discoveries is small, so is the probability forintermediate nodes to knowan active route to a destination node. Consequently, a large number ofAODV route requests (RREQ) must travel up to the destination node. However, as the data rateincreases, so does the chance for intermediate nodes to have cached active routes, while OLSRmust completely reconfigure its routing tables. Accordingly, there is a threshold below whichOLSR is beneficial for VANET, and above which AODV becomes attractive. From Fig. 4, thisthreshold is situated around 0:8Mbits=s. The Routing Overhead (RO) is depicted in Fig. 5. It revives the old cleavage betweenproactive and reactive routing protocols. Indeed, reactive protocols have been initially developedto reduce the routing overheads created by the proactive approaches. However, this assumptionwas shown to be valid if the traffic rate, and so the route discovery rate, Fig. 4. Packet Delivery Ratio (PDR) as a function of Data Traffic Ratewas not too large. Above a certain traffic rate threshold, itwas assumed that table-drivenapproaches were more attractive than on-demand schemes. In Fig. 5, we actually see that thiscleavage also exists in VANETs. We observe that the control traffic of OLSR exhibits theexpected characteristics of being independent of the data traffic rate, while the control traffic,generated by AODV, increases with the data rate. For data rate below 300kbit=s, AODV has alower routing overhead than OLSR, while for data rate above this threshold, the control trafficgenerated by AODV explodes. 65
  9. 9. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEME We observe in Fig. 6 that OLSR consistently presents the lowest delay, regardless ofdata traffic. This may be explained by the fact that OLSR, as a proactive protocol, has a fasterprocessing at intermediate nodes. When a packet arrives at anode, it can immediately beforwarded or dropped. In reactive protocols, if there is no route to a destination, packets tothat destination will be stored in a buffer while a route discoveryis conducted. Accordingly,the performance improvement interm of delay raises up to 250% between AODV and OLSR. Finally, we show in Fig. 7 the expected number of hops of the CBR routes, whichreflects the average end-to-endroute length. Three remarks may be made on this figure. First,the maximum average number of hops never goes beyond 2 hops. According to thesimulation area and the transmission range, it should be situated between 3 and 4hops. Bylooking at Fig. 4, we see that the number of hops is not related to the data rate, as we have 2hops and 95% PDR at low rate. The only explanation comes from the non-uniformdistribution of vehicles’ in the simulation environment. Indeed, vehicles are aggregating atintersections, and the intersections are globally located toward the center of the simulationenvironment. Accordingly, the effect increases the connectivity at the intersection andbetween intersections, and consequently lowers the number of hops. Second, we see that thenumber of hops of AODV is always larger than the number of hops of OLSR. As themaximum number of hops is approximately 2 hops on average, and as MPR has beenpurposely created to optimize the number of hops in its two hops neighborhood, it is notsurprising that AODV creates routes 15% longer than OLSR. The last remark is that the pathlength actually decreases as the CBR rate increases. Thisis also not a surprise since anincreased number of hops also increases the probability to loose packets. So, as the networkbecomes saturated, only packets with the shortest path may be correctly received. This is,unfortunately, a major illustration of network unfairness toward traffic flows.In the next set of figures, we display results obtained for the second scenario. Node density isdefined as a node’s average number of neighbors and is computed as mentioned in Table I.Fig. 8 shows the typical bell shape of AODV and OLSR’s PDRs. For small densities, thereare not enough nodes to ensure network connectivity. So, as we increase 66
  10. 10. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEMEthe density and leave the supra-critical 1 zone, the PDR gets improved until the density of nodesreaches the critical value. Then, as the density still increases, we drop into asuper-critical1 zone,where extra nodes are able to provide some redundancy for route management. As neither OLSR norAODV are configured to benefit from load balancing in our implementation, the extra number ofnodes soon becomes a drawback for the MAC layer. Consequently, the PDR starts dropping. Thecritical density in our simulation is between 4 and 6 nbrs/vhcl on average. Although this situation iscommon in MANETs, it is worsen by the non uniform distribution of nodes in the simulation area.Indeed, due to traffic regulations and vehicles configurations, urban traffic tends to cluster atintersections, which locally increases the density and decreases the performance of VANET routingprotocols. The interesting part in Fig 8 is that AODV and OLSR are sustaining the clustering effectdifferently. For low density, OLSR outperforms AODV. Then, similarly to Fig. 4, a threshold isreached, as the density increases, above which AODV starts outperforming OLSR. In order to analyzethis graph, we separate the graph in three regions: supra-critical, critical, and super-critical densities.In the supra-critical density (4 nbrs/vhcl and below), OLSR performs better than AODV, which is anote worthy effect here as OLSR seems to handle network disconnection betterthan AODV. Networkdisconnections are an unwanted, yet common, problem in VANETs. It therefore seems that OLSRcould be a good candidate for routing in sparse VANET networks. Then, in the critical category,OLSR still maintainsits advantage toward AODV. Indeed, when cars are aggregating in intersections,the MPR nodes become more stable, which increases the stability of OLSR and helps improving thePDR. Finally, in the super-critical category, the PDR for both protocols drops. However, the PDR ofOLSR drops faster than AODV, which seems to be handling saturated networksbetter than OLSR.One explanation must again be soughtin the intersection. As the density of car locally increases, theperiodic maintenance of OLSR reduces its capability ofaccessing the channel for data traffic, whilethe AODV’s RREQ packets have a high chance to find a close intermediate node with an open route. 67
  11. 11. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEME Similarly to Fig. 5, the next figure depicts the RO of OLSR and AODV as a function ofthe node density. We can see on Fig. 9 that as we could expect, both ROs increase with thedensity. As in Fig. 5, we clearly see a transition threshold for the control traffic generated byOLSR and AODV. For node densities below 8 nbrs/vhcl, the control traffic overhead ofAODV is smaller than OLSR. However, as the density increases, the cost of repeated routediscovery procedures in AODV introduces a large control traffic overhead and OLSRends upoutperforming AODV up to 100%. Figure 10 depicts the end-to-end packet delay. For bothsupra-critical and criticaldensities, both protocols have similar delays. However, in the super-critical zone, AODV’sdelay explodes and, as the confidence interval are illustrating, it isalso more unstable. As theaccesses to the channel become shared, when a RREQ finds an intermediate node withanactive route the delay can be lowered. However, the penalty fornot finding anyintermediate node becomes prohibitive as the network becomes locally saturated. On theother hand, routes that OLSR could maintain despite the congested channel are ready to use.So, we have an ambiguous result here, where insaturated networks, OLSR has a lower PDRthan AODV, yet the packets that it manages to carry are delivered with a much smaller delay. 68
  12. 12. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEME Finally, Fig. 11 shows the expected number of hops ofthe CBR routes as a function ofthe density of vehicles. Similar remarks may be formulated as for Fig. 7 on the maximumnumber of hops and on the 15% increase in the number of hops of AODV compared toOLSR. Yet, the average number of hops’ behavior toward the density of vehicles is slightlydifferent. Indeed, Fig. 11 has a typical bell shape. As the density of vehicles increases, sodoes the path length. By looking at Fig. 8, we see that the PDR is similarly increasing.Accordingly, unlike Fig. 7, network disconnections due to a low vehicle density arerestricting multi-hop communications. Then, the density increases as the length of multi-hoproutes. However, after a certain threshold, the network becomes saturated and a similar effectcan be observed as with the increased data rate. The conclusion from this is that, similarly toMANETs, reliable multi-hop communication may only occur in a particular threshold, wherethe network density is large enough to be connected, yet moderate enough in order to limitthe channel saturation. But the situation is worsening in VANETs by the clustering effect atthe intersections, as the density might be too large to keep reliable single hopcommunications, yet too low to maintain multi-hop communications.6. CONCLUSION AND FUTURE WORKS In this paper, we evaluated the performance of AODV and OLSR for vehicular ad hocnetworks in urban environments. The traffic regulations and the vehicles characteristicshandled by the vehicular mobility model (VMM) we used are creating a clustering effect atintersection. This effect has a remarkable property on standard performance evaluations of adhoc protocols. The first one is that neither the initial nor the maximum velocity has anyinfluence on routing protocols in urban environments. Indeed, due to the interaction with thespatial environment and also other neighboring cars, vehicles experience non negligible speeddecay independent of the network velocity. Then, a second property is local increase of nodesdensity, which clearly has a consequence on both testedad hoc routing protocols.We testedOLSR and AODV against node density and data traffic rate. Globally, we found that OLSRoutperforms AODV in VANETs. For most of the metrics we used in this paper, OLSR hasbetter performance that AODV. Indeed, OLSR has smaller routing overhead, end-to-enddelay and route lengths. And for the PDR, where OLSR may be outperformed by AODVafter a certain threshold, the performance loss is limited to 10%. Accordingly, unlike aprevious study for MANET([7]), which suggested that neither OLSR nor AODV couldoutperform each others, or even all previous studies described in Section II, OLSR, aproactive protocol, is more fitted to VANET than reactive ones. We also illustrated in thispaper how vehicular ad hoc networks in urban environment experience particular motionpatterns. More precisely, we showed that the average velocity was not a valid parameter to 69
  13. 13. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEMEevaluate routing protocolsin VANET under realistic motion patterns. Accordingly, for futurerealistic performance evaluation, one should rather evaluate ad hoc protocols against newmetrics, such as acceleration/deceleration capabilities of the drivers, or the length of streetsegments instead of simple average mobility. For this study, we deliberately parameterizedthe network to be fully connected, as we wanted to avoid biased resultsfrom disconnectedgraphs. However, as stated in the paper, network disconnections are also a major property ofVANETs and we will perform similar tests with shorter transmission range. We are alsointerested in evaluating the effect of heterogeneous vehicles in urban environments on routingprotocols for VANETs. Finally, we plan to include geo graphical forwarding protocols infuture performance evaluation as they are more suited to dense networks or to frequentnetwork disconnections.REFERENCES[1] R. Akcelik, M. Besley, “Acceleration and Deceleration Models”, 23rdConference of Australian Institutes of Transport Research (CAITR2001), Melbourne, Australia, December 2001.[2] J. Broch etal. “A Performance Comparison of Multi-Hop Wireless AdHoc Network Routing Protocols”, In Proc. ACM MOBICOM 98, Dallas, TX, October 1998.[3] T.H. Clausen and P. Jacquet, “Optimized Link State Routing (OLSR)”, RFC 3626, October 2003.[4] T. Camp, J. Boleng, V. Davies, ”A Survey of Mobility Models for Ad Noc Network Research”, Wireless Communications and Mobile Computing, Vol. 2, Issue 5, pp. 483- 502, September 2002.[5] S. R. Das et al., ”Comparative Performance Evaluation of Routing Protocols for Mobile Ad-Hoc Networks”, In 7th Int. Conf. on Comp. Communication and Networks, pp. 153- 161, Lafayette, LA, Oct. 1998.[6] J. Haerri, F. Filali, C. Bonnet, “A Framework for Mobility Models Generation and its Application to Inter-Vehicular Networks”, 3rd IEEE International Workshop on Mobility Management and Wireless Access (MobiWac’05), Maui, Hawaii, U.S.A., June 2005,[7] D. Helbing, “Traffic and Related Self-driven Many-particles Systems”, Rev. Modern Physics, Vol. 73, pp. 1067-1141, 2001.[8] Jerome Haerri, Marco Fiore, Fethi Filali, Christian Bonnet, Carla-Fabiana Chiasserini, Claudio Casetti, ”A Realistic Mobility Simulatorfor Vehicular Ad Hoc Networks”, Eur´ecom Technical Report, Institut Eur´ecom, France, 2005.[9] P. Johansson et al. , ”Scenario-based Performance Analysis of Routing Protocols for Mobile Ad-Hoc Networks”, In Proc. IEEE/ACM Mobicom’99, pp. 195-206, Seattle, WA, Aug. 1999.[10] A. Jardosh, E. Belding-Royer, et al., ”Toward realistic mobility models for mobile ad hoc networks”, in Proc. of the 9th Annual International Conference on Mobile Computing and Networking (MobiCom 2003),September 2003.[11] C. Perkins, E. Belding-Royer, S. Das, quet, ”Ad hoc On-Demand Distance Vector (AODV) Routing”, RFC 3561, July 2003.[12] R.A. Santos et al.,”Performance Evaluation of Routing Protocols in Vehicular Ad Hoc Networks”, in International Journal of Ad Hoc and Ubiquitous Computing 2005, Vol. 1, No.1/2, pp. 80 - 91.[13] Sven Jaap, Marc Bechler, and Lars Wolf, ”Evaluation of Routing Protocols for Vehicular Ad Hoc Networks in City Traffic Scenarios”, in Proc of the 5th International Conference on Intelligent Transportation Systems Telecommunications (ITST), Brest, France, June 2005. 70
  14. 14. International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN0976 – 6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 2, March – April (2013), © IAEME[14] Sven Jaap, Marc Bechler, and Lars Wolf, ”Evaluation of Routing Protocols for Vehicular Ad Hoc Networks in City Traffic Scenarios”, in Proc of the 5th International Conference on Intelligent Transportation Systems Telecommunications (ITST), Brest, France, June 2005.[15] Thomas Heide Clausen, Philippe Jacquet, Laurent Viennot, “Comparative Study of Routing Protocols for Mobile Ad-hoc Networks”, in 1stIFIP MedHocNet Conference, 2002.[16] M. Treiber, A. Hennecke, D. Helbing, “Congested traffic states inempirical observations and microscopic simulations”, Phys. Rev. E 62, Issue 2, August 2000.[17] M. Treiber, D. Helbing, “Realistische Mikro simulation von Strassenverkehrmit einem einfachen Modell”, 16th Symposium Simulation stechnik ASIM 2002, Rostock, September 2002.[18] The Fleet Net Project,[19] The Network on Wheels Project,[20] T.Priyadarsini, B.Arunkumar, K.Sathish and V.Karthika, “Traffic InformationDissemination in Vanet using IEEE-802.11” International journal of Electronics andCommunication Engineering &Technology (IJECET), Volume 4, Issue 1, 2013,pp. 294 - 303, ISSN Print: 0976- 6464, ISSN Online: 0976 –6472.[21] S. A. Nagtilak and U.A. Mande, “The Detection of Routing Misbehavior in Mobile AdHoc Networks Using the 2ack Scheme With OLSR Protocol” International journal ofComputer Engineering & Technology (IJCET), Volume 1, Issue 1, 2010, pp. 213 - 234,ISSN Print: 0976 – 6367, ISSN Online: 0976 – 6375.[22] Sunita Kushwaha, Bhavna Narain, Deepti Verma and Sanjay kumar, “Effect of ScenarioEnvironment on the Performance of Manets Routing Protocol: AODV” International journalof Computer Engineering & Technology (IJCET), Volume 2, Issue 1, 2011, pp. 33 - 38, ISSNPrint: 0976 – 6367, ISSN Online: 0976 – 6375.[23] M. Pushpalatha, T. Ramarao, Revathi Venkataraman and Sorna Lakshmi, “MobilityAware Data Replication using Minimum Dominating Set in Mobile Ad Hoc Networks”International journal of Computer Engineering & Technology (IJCET), Volume 3, Issue 2,2012, pp. 645 - 658, ISSN Print: 0976 – 6367, ISSN Online: 0976 – 6375. 71